Resumo: 1. As armadilhas convencionais são amplamente utilizadas em estudos com pequenos mamíferos não-voadores, mas geralmente acarretam um viés de amostragem. Além disso, efeitos da qualidade do microhabitat e da sazonalidade podem influenciar a estimativa da diversidade. 2. Nós comparamos dois métodos de amostragem de pequenos mamíferos: armadilhas de captura viva e de queda, sendo a riqueza de espécies e número de indivíduos variáveis resposta. Os levantamentos de campo foram realizados em uma área contínua de Mata Atlântica. Os efeitos das variáveis de microhabitat e da sazonalidade sobre capturabilidade também foram investigados. 3. As armadilhas de queda capturaram maior número de espécies e indivíduos durante a estação chuvosa, quando comparadas com armadilhas de captura viva. Portanto, as armadilhas de queda foram o melhor método para estimar a riqueza e abundância de pequenos mamíferos, principalmente de espécies semi-fossoriais. 4. Nossos resultados mostraram a importância de considerar o esforço amostral despendido em armadilhas de captura vida e de queda ao se comparar a eficiência de captura entre elas. O uso da abundância relativa resolveu problemas de sub e de superestimativa da capturabilidade em cada tipo de armadilha. 5. Embora o microhabitat seja considerado um importante fator que pode influenciar o número de indivíduos em determinada área de estudo, não encontramos influência do mesmo na capturabilidade de armadilhas de captura viva e de queda na comunidade de pequenos mamíferos. No entanto, verificamos que as estações de captura que apresentaram maior área basal de samambaias e de árvores mortas capturaram mais indivíduos de Brucepattersonius soricinus, Oligoryzomys nigripes e Akodon montensis. 6. Concluímos, portanto que a maioria dos inventários de pequenos mamíferos em regiões tropicais não conseguiu captar a riqueza... (Resumo completo, clicar acesso eletrônico abaixo) / Abstract: 1. Conventional traps are widely used in studies of non-volant small mammals, but usually lead to sampling bias. Also, microhabitat quality and seasonality effects may influence the estimate of diversity. 2. We compared two small mammal sampling methods: live and pitfall traps. Species richness and number of individuals were our response variables, and field surveys were carried out in a continuous area of the Brazilian Atlantic Forest. Effects of microhabitat variables and seasonality on trappability were also investigated. 3. Pitfall traps captured more number of species and more individuals during the wet season, when compared with live traps. Therefore, pitfall-traps were a better method for estimating the richness and abundance of small mammals, especially semi-fossorial species. 4. Our results showed the importance of considering the sampling effort dispended in live and pitfall traps when comparing the capture efficiency between them. The use of weighted abundance solved trappability problems of underestimation and overestimation in each trap type to capture small mammals. 5. Although the microhabitat is considered an important factor that can influence the number of individuals in the study area, we found no influence of microhabitat on trappability of live and pitfall traps on small mammal community. However, we found that the trap stations with higher basal area of dead trees and ferns captured more individuals of Brucepattersonius soricinus, Oligoryzomys nigripes, and Akodon montensis. 6. We conclude that most small mammal inventories in tropical regions failed to capture the true species richness. Several species considered "rare" and underlisted as data deficient in IUCN criterion could be in fact common / Orientador: Mauro Galetti / Coorientador: Maria José de Jesus Silva / Banca: Marco Aurélio Pizo Ferreira / Banca: Denis Cristiano Briani / Mestre

Made available in DSpace on 2014-06-11T19:30:14Z (GMT). No. of bitstreams: 0 Previous issue date: 2010-12-02Bitstream added on 2014-06-13T18:59:59Z : No. of bitstreams: 1 neves_cl_me_rcla.pdf: 579299 bytes, checksum: 4e0eb889aafa2deb8e85b3bfdbcfb15a (MD5) / Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq) / As armadilhas convencionais são amplamente utilizadas em estudos com pequenos mamíferos não-voadores, mas geralmente acarretam um viés de amostragem. Além disso, efeitos da qualidade do microhabitat e da sazonalidade podem influenciar a estimativa da diversidade. 2. Nós comparamos dois métodos de amostragem de pequenos mamíferos: armadilhas de captura viva e de queda, sendo a riqueza de espécies e número de indivíduos variáveis resposta. Os levantamentos de campo foram realizados em uma área contínua de Mata Atlântica. Os efeitos das variáveis de microhabitat e da sazonalidade sobre capturabilidade também foram investigados. 3. As armadilhas de queda capturaram maior número de espécies e indivíduos durante a estação chuvosa, quando comparadas com armadilhas de captura viva. Portanto, as armadilhas de queda foram o melhor método para estimar a riqueza e abundância de pequenos mamíferos, principalmente de espécies semi-fossoriais. 4. Nossos resultados mostraram a importância de considerar o esforço amostral despendido em armadilhas de captura vida e de queda ao se comparar a eficiência de captura entre elas. O uso da abundância relativa resolveu problemas de sub e de superestimativa da capturabilidade em cada tipo de armadilha. 5. Embora o microhabitat seja considerado um importante fator que pode influenciar o número de indivíduos em determinada área de estudo, não encontramos influência do mesmo na capturabilidade de armadilhas de captura viva e de queda na comunidade de pequenos mamíferos. No entanto, verificamos que as estações de captura que apresentaram maior área basal de samambaias e de árvores mortas capturaram mais indivíduos de Brucepattersonius soricinus, Oligoryzomys nigripes e Akodon montensis. 6. Concluímos, portanto que a maioria dos inventários de pequenos mamíferos em regiões tropicais não conseguiu captar a riqueza... / 1onventional traps are widely used in studies of non-volant small mammals, but usually lead to sampling bias. Also, microhabitat quality and seasonality effects may influence the estimate of diversity. 2. We compared two small mammal sampling methods: live and pitfall traps. Species richness and number of individuals were our response variables, and field surveys were carried out in a continuous area of the Brazilian Atlantic Forest. Effects of microhabitat variables and seasonality on trappability were also investigated. 3. Pitfall traps captured more number of species and more individuals during the wet season, when compared with live traps. Therefore, pitfall-traps were a better method for estimating the richness and abundance of small mammals, especially semi-fossorial species. 4. Our results showed the importance of considering the sampling effort dispended in live and pitfall traps when comparing the capture efficiency between them. The use of weighted abundance solved trappability problems of underestimation and overestimation in each trap type to capture small mammals. 5. Although the microhabitat is considered an important factor that can influence the number of individuals in the study area, we found no influence of microhabitat on trappability of live and pitfall traps on small mammal community. However, we found that the trap stations with higher basal area of dead trees and ferns captured more individuals of Brucepattersonius soricinus, Oligoryzomys nigripes, and Akodon montensis. 6. We conclude that most small mammal inventories in tropical regions failed to capture the true species richness. Several species considered “rare” and underlisted as data deficient in IUCN criterion could be in fact common

The Amazon rainforest harbors the planet's highest biodiversity among terrestrial ecosystems; however, the biology and ecology of most of its species are unknown. Niche partitioning is considered a key factor allowing species co-existence, especially for morphological similar species such as spiny rats of the genus Proechimys. I examined the extent to which habitat differentiation, species body mass, and diet could explain the community composition of small mammals at a site in the southeastern Amazon. Moreover, I radio-tagged sympatric species of Proechimys spp. and Mesomys stimulax (an arboreal spiny rat) to obtain detailed autoecology information, including habitat use and use of space. I found support for niche partitioning among species and associated small mammal species with distinct successional phases of gap dynamics. I also observed among Proechimys spp. a typical polygynous organization: females appeared to be territorial against females of any species of their genus; but male territories overlapped with those of several females. Mesomys stimulax showed evidence of monogamy and possible sociality, although sample sizes were small. To identify the importance of small mammals as seed disperser of Brazil nut seeds, I conduct experiments using a combination of fluorescent powder, seed exclosures, and track plates in forests with and without Brazil nut groves, and in forests with and without Brazil nut harvesting. Among small mammals, only Proechimys spp. removed, dispersed, and preyed upon Brazil nut seeds. Proechimys spp. sometimes scatterhoarded these seeds, and hence have the potential to play a significant role in recruiting new Brazil nut trees. I also trapped small mammals and measured forest structures on the same sites used for the seed dispersal experiment, to determine the ecological effects of Brazil nut harvest on small mammal communities; however, my results showed little evidence of changes associated with the seed exploitation. My results highlight the importance of habitat heterogeneity in structuring small mammal communities, and indicate that forest management practices that alter habitats, such as partial logging, also can be expected to alter small mammal composition and diversity. Proechimys spp. have the potential to play an important role in the ecological restoration of intensive exploited Brazil nut groves.

The Amazon rainforest harbors the planet's highest biodiversity among terrestrial ecosystems; however, the biology and ecology of most of its species are unknown. Niche partitioning is considered a key factor allowing species co-existence, especially for morphological similar species such as spiny rats of the genus Proechimys. I examined the extent to which habitat differentiation, species body mass, and diet could explain the community composition of small mammals at a site in the southeastern Amazon. Moreover, I radio-tagged sympatric species of Proechimys spp. and Mesomys stimulax (an arboreal spiny rat) to obtain detailed autoecology information, including habitat use and use of space. I found support for niche partitioning among species and associated small mammal species with distinct successional phases of gap dynamics. I also observed among Proechimys spp. a typical polygynous organization: females appeared to be territorial against females of any species of their genus; but male territories overlapped with those of several females. Mesomys stimulax showed evidence of monogamy and possible sociality, although sample sizes were small. To identify the importance of small mammals as seed disperser of Brazil nut seeds, I conduct experiments using a combination of fluorescent powder, seed exclosures, and track plates in forests with and without Brazil nut groves, and in forests with and without Brazil nut harvesting. Among small mammals, only Proechimys spp. removed, dispersed, and preyed upon Brazil nut seeds. Proechimys spp. sometimes scatterhoarded these seeds, and hence have the potential to play a significant role in recruiting new Brazil nut trees. I also trapped small mammals and measured forest structures on the same sites used for the seed dispersal experiment, to determine the ecological effects of Brazil nut harvest on small mammal communities; however, my results showed little evidence of changes associated with the seed exploitation. My results highlight the importance of habitat heterogeneity in structuring small mammal communities, and indicate that forest management practices that alter habitats, such as partial logging, also can be expected to alter small mammal composition and diversity. Proechimys spp. have the potential to play an important role in the ecological restoration of intensive exploited Brazil nut groves.